Freight Forwarder’s Pricing Strategy Incorporating Learning from Repeated Transactions

As logistics service providers, freight forwarders are intermediary parties which connect cargo owners to carriers in the logistics chain. Pricing strategies are the approaches a freight forwarder could use to determine the price of its services. The purpose of this research is to help freight forwarders design and implement their pricing strategy by incorporating the effect of learning from past transactions. A multi-agent framework is built to investigate the interaction between cargo owners, freight forwarders and carriers. Simulations are then conducted to investigate the price performance of freight forwarders over repeated transactions. Freight forwarders can learn either by reinforcement learning or on an “if-then” basis to adapt their pricing decisions. The experiments with the simulations show that freight forwarders using reinforcement learning outperform those using non-learning pricing strategies, but are in turn outperformed by freight forwarders which learn on an “if-then” basis.